Abstract
Robot-assisted kidney transplantation (RAKT) has emerged as a minimally invasive alternative to open kidney transplantation (OKT). However, the comparative benefits and limitations of both approaches remain unclear. To compare the outcomes of RAKT and OKT through an umbrella review of published systematic reviews and meta-analyses. This umbrella review was written following PRISMA guidelines. Only meta-analysis and systematic review were included. Many outcomes were studied such as warm (WIT) and cold ischemia times (CIT), rewarming time, total ischemia time, blood loss, blood transfusion, delayed graft function, surgical site infections (SSI), hospital stay, graft rejection, graft failure, all-cause mortality, operation time, incision length, hospital readmission, vascular and ureteral anastomosis time, and other complications. We assessed methodological quality via the AMSTAR-2 tool. Consistency and strength of evidence were also evaluated. 6 studies were included: four meta-analyses and two systematic reviews. Studies were of low to moderate quality. Robotic kidney transplant surgery yields better outcomes in terms of blood loss (MDs ranging from − 16 to − 55 mL across studies), post-operative pain, incision length, SSI and ureteral leak. On the other hand, CIT, rewarming time, total ischemia time, operative time and ileus rate were favored in open surgery. Same rejection rates, graft failure, mortality and hospital readmission rate were found between both techniques. RAKT appears to improve several perioperative outcomes while maintaining comparable graft and patient survival outcomes to OKT. However, the available evidence remains limited by the low-to-moderate quality of existing reviews, highlighting the need for high-quality prospective studies.
Keywords: Robotic kidney transplant, Open kidney transplant, Peri-operative outcomes, Complications, Umbrella study
Introduction
Renal transplant is the most applied transplant surgery in the world [1]. Introduced in 1954 in Boston, Massachusetts [2], it made proof of its efficiency in term of quality of life compared to dialysis [3]. Moreover, with the better understanding of immunosuppressor and specialized transplant teams, success rate of renal transplant has reached around 97.8% [4]. End-Stage Renal Disease (ESRD) is a highly morbid disease with an incidence of around 7% [5]. Aside from its morbidity, ESRD constitutes a very high economic burden on society, representing around 6.4% of the GDP in America [5]. Accordingly, kidney transplant could constitute an optimal solution for patients with ESRD.
Kidney transplant is a complex surgery consisting of three main steps: arterial, venous and ureteral anastomosis. Hence, surgical mastery is of the essence for the success of the surgery. Therefore, technique improvements are essential to avoid the loss of the transplanted kidney, especially given the scarcity of donors. Open kidney transplant (OKT) is the standard approach, with great success, despite its disadvantages such as large incisions, increased pain and long recovery times.
With the emergence of robotic surgery, we are progressively watching a surgical shift from traditional open techniques to minimally invasive robot-assisted laparoscopy. The latter provides a comfortable setting for surgeons to work in [6] which can affect the whole surgery. Aside from that, it provides the surgeon with enhanced precision and a close-up view of the surgical site. Other reported benefits in the literature include a smaller incision, lower bleeding risk, shorter hospital stay, and lower surgical site infection rates, ….
The first robot-assisted kidney transplant (RAKT) surgery was performed in 2002 in France, using the Da Vinci robot with favorable outcomes [7]. That said, robotic surgery has found its way into transplant surgery, becoming more and more used by surgeons all around the world [8].
Most recently, Afferi et al. published two large multicenter propensity score-matched analyses comparing RAKT to OKT in both living and deceased donor settings across seven European academic centers, representing the highest level of non-randomized comparative evidence currently available and providing important donor-specific insights into the relative merits of each approach [9, 10].
Many studies compared the outcome of RAKT and OKT, of which randomized trials, systematic reviews and meta-analysis. However, some studies have inconsistent finding, vary in methodological quality and evaluate different outcomes.
An umbrella review, also called an overview of reviews, is the study of multiple meta-analysis and systematic reviews and are considered the highest level of evidence in medicine [11].
Therefore, we decided to conduct an umbrella review. The aim of this study is to compare the outcome of kidney transplants between open and robotic techniques in terms of surgical complications, surgical and revascularization time, wound infection, post-operative pain, speed of recovery and transplant outcome.
In other words, we aim to answer the following question: How do the surgical outcomes of robot-assisted kidney transplantation compare to those of open kidney transplantation in terms of surgical complications, operative efficiency, post-operative recovery, and overall morbidity?
Methods
This study is an umbrella review comparing the outcomes of robotic versus open renal transplant. It was conducted according to the PRISMA guidelines.
The outcomes studied include warm (WIT) and cold ischemia times (CIT), rewarming time, total ischemia time, blood loss, blood transfusion, delayed graft function, surgical site infections (SSI), hospital stay, graft rejection, graft failure, all-cause mortality, operation time, incision length, incisional hernia, lymphocele, venous thrombosis, arterial stenosis, ureter leak, urinary tract infection, post-operative pain, ileus rate, hospital readmission, arterial anastomosis time, venous anastomosis time and ureteral anastomosis time.
A comprehensive search of PubMed, Scopus, and Web of Science was conducted in late 2024. The following search strategy was applied in PubMed: (((“robotic“[tiab] OR “robot-assisted“[tiab] OR “robot assisted“[tiab] OR “RAKT“[tiab] OR “minimally invasive“[tiab]) AND (“kidney transplant*“[tiab] OR “renal transplant*“[tiab] OR “kidney transplantation“[MeSH] OR “renal transplantation“[MeSH])) AND ((“open“[tiab] OR “conventional“[tiab] OR “standard“[tiab]) AND (“kidney transplant*“[tiab] OR “renal transplant*“[tiab]))) AND (“systematic review“[tiab] OR “meta-analysis“[tiab] OR “meta analysis“[tiab] OR “systematic review“[pt] OR “meta-analysis“[pt]).
The following search strings were adapted for Scopus and Web of Science: ((“robotic” OR “robot-assisted” OR “RAKT” OR “minimally invasive”) AND (“kidney transplant*” OR “renal transplantation”) AND (“open” OR “conventional”) AND (“systematic review” OR “meta-analysis”)). No date or language restrictions were applied.
Studies reporting any surgical outcome were included in this review, regardless of country and publication date. Only meta-analysis and systematic reviews were included in our study. All case reports, narrative reviews and editorials were excluded. Study selection and data extraction were performed independently by both authors (R.M. and E.H.); any disagreements were discussed and resolved by consensus, with no cases requiring adjudication by a third party.
From each article data extracted included study type, name of primary author, year of publication, the journal in which the study was published, the number of subjects and/or studies included and the parameters studies. After selection of studies, all the outcomes were extracted with the effect sizes (e.g., risk ratios [RR], mean differences [MD], or standardized mean differences [Hedges’ g]), the confidence intervals (CI) and the measures of heterogeneity (I²) (If present).
The methodological quality of each included review was assessed using the AMSTAR-2 tool, a validated instrument for appraising systematic reviews of healthcare interventions. Each review was rated as high, moderate, low, or critically low quality based on its adherence to 16 domains (Fig. 1).
Fig. 1.
AMSTAR-2 validation tool
Consistency of findings across reviews was evaluated qualitatively. Outcomes were categorized as:
Consistent (≥ 75% of studies reporting same direction of effect).
Partially consistent (50–74%).
Inconsistent (< 50%).
The overall strength of evidence was graded by combining consistency and AMSTAR-2 quality ratings (That included 7 critical AMSTAR-2 items) (Table 1) for each outcome.
Table 1.
AMSTAR-2 quality rating
| Quality of AMSTAR criteria | |
|---|---|
| There are 7 critical items | |
| AMSTAR-2 Item # | Description |
| 2 | Protocol registered |
| 4 | Comprehensive search strategy |
| 7 | Excluded studies listed with reasons |
| 9 | Risk of bias assessed for included studies |
| 11 | Appropriate statistical methods |
| 13 | Heterogeneity explained |
| 15 | Conflicts of interest disclosed |
| Rating | Criteria |
| □ High | No or 1 non-critical weakness |
| □ Moderate | More than 1 non-critical weakness, no critical flaws |
| □ Low | 1 critical flaw, with or without non-critical weaknesses |
| □ Critically Low | > 1 critical flaw or major issues in protocol, methods, or bias handling |
A formal overlap assessment was performed using the Corrected Covered Area (CCA) methodology as described by Pieper et al. Across the 6 included reviews and 32 unique primary studies, a total of 71 study inclusions were identified, yielding a CCA of 24.4%, indicating very high overlap. Several primary studies — notably Kishore 2020, Maheshwari 2020, Pein 2020, Tugcu 2018, and Oberholzer 2013 — each appeared in five of the six included reviews. Findings were therefore interpreted based on consistency of effect direction and AMSTAR-2 quality ratings rather than frequency of reporting, to avoid overweighting results driven by repeated inclusion of the same primary data.
This review followed the PRISMA 2020 guidelines for umbrella reviews. A PRISMA flow diagram was used to illustrate the study selection process.
Artificial intelligence (AI)-assisted language editing tools were used solely to improve grammar, sentence structure, and wording. The scientific content, study design, data interpretation, results, and conclusions were developed and reviewed by the authors, who take full responsibility for the content of the manuscript.
This study being a review, ethics approval was not necessary.
Results
Initial PubMed, Scopus, and Web of Science search found 51 articles. After title, abstract and full text screening, only meta-analysis and systematic reviews were included, and a total of 6 studies were included: Four meta-analysis and two systematic reviews.
The study selection process is outlined in the PRISMA flow diagram (Fig. 2).
Fig. 2.
PRISMA flow diagram
The reviews included compared RAKT and OKT. They were published from 2017 to 2023 and included 6 to 18 primary studies. Each study had different reported outcome. A summary of study characteristics is presented in Table 2.
Table 2.
Study characteristics
| Title | Type | Author | Country | Journal | Year | Number of subjects | Goal | Parameters |
|---|---|---|---|---|---|---|---|---|
| Article 1 | Meta-Analysis | Guangxiang Lui | China | BioMed Research International | 2020 |
804 + 263 (6 studies) |
Compare the safety and efficacy of RAKT versus OKT. | Age, sex, BMI, IS treatment, FU duration, type of donors, WIT, CIT, rewarming time, total ischemia time, blood loss, blood transfusion, delayed graft function, SSI, length of hospital stay, graft rejection, graft failure, all cause mortality |
| Article 2 | Meta-Analysis | Julian Slagter | Netherlands | International journal of surgery | 2022 |
1316 + 482 (11 studies) |
Compare the clinical outcomes of RAKT with open kidney transplantation | Age, BMI, sex, donor type, pre-emptive, dialysis time, Op time, console time, CIT, Rewarming time, blood loss, incision length, type of incision, conversion to open, cooling system, IS, SSI, incisional hernia, lymphocele, pain score, DGF, acute rejection, hosp stay, cost, creatinine, arterial stenosis, venous thrombosis, ureter leakage, |
| Article 3 | Meta-Analysis | Kumar Madhavan | India | Indian Journal of Urology | 2023 |
1878 + 677 (16 studies) |
Compare the clinical outcomes of robot-assisted kidney transplant (RAKT) to open kidney transplant (OKT) | Age, gender, BMI, donor type, dialysis, cooling method, IS, op time, anastomosis time (arterial, venous and ureteral), WIT, CIT, blood loss, incision length, Pain score, SSI, Incisional hernia, Lymphocele, graft thrombosis and stenosis and ureter-related complications, DGF, UTI, hosp stay, Acute rejection, Creatinine, graft survival and OS |
| Article 4 | Meta-Analysis | Mario A O’Connor-Cordova | Mexico | Transplantation reviews | 2023 |
969 + 508 (9 studies) |
Primary outcomes of interest were surgical times, ischemia times, blood loss, complication rates, and graft function | Ischemia time, rewarming time, procedure time, time to complete anastomosis, blood loss, transfusion rate, SSI, lymphocele, ileus rate, pain scores, length of hospital stay, hospital readmission rates, DGF, acute rejection, early graft outcomes |
| Article 5 | Systematic Review | Sven Wagenaar | Netherlands | European Urology | 2017 | 18 studies | Evaluate, summarize, and review evidence supporting operating technique and postoperative outcome for kidney transplant recipients | BMI, op time, ischemia time, death, post-op kidney function, wound complications, hernia, incision length, post-op time, recovery time |
| Article 6 | Systematic Review | A Territo | Spain | Actas Urologicas Espanolas | 2023 |
1552 + 584 (13 studies) |
Gather comparative data of OKT vs. RAKT in a systematic review | Pre-emptive rate, op time, complication rates, DGF, graft survival |
Article 1: Robot-Assisted versus Conventional Open Kidney Transplantation: A Meta-Analysis
Article 2: Robot-assisted Kidney transplantation as a minimally invasive approach for kidney transplant recipients: A systematic review and meta-analyses
Article 3: Comparison of outcomes after open versus robotic Kidney transplantation: A systematic review and meta-analysis
Article 4: Living donor robotic assisted kidney transplant compared to traditional living donor open kidney transplant. Where do we stand now? A systematic review and Meta-analysis
Article 5: Minimally Invasive, Laparoscopic, and Robotic-assisted Techniques Versus Open Techniques for Kidney Transplant Recipients: A Systematic Review
Article 6: Open versus robotic-assisted kidney transplantation: A systematic review by the European Association of Urology (EAU) - Young Academic Urologists (YAU) Kidney Transplant Working Group
WIT: Warm ischemia time; CIT: Cold ischemia time; SSI: Surgical site infections; UTI: Urinary tract infection
Using the AMSTAR-2 tool, two reviews were rated as low, two were moderate to low and two were of moderate rating. No reviews were of high rating.
AMSTAR-2 rating and key strength and weakness for each article are shown in Table 3.
Table 3.
AMSTAR-2 rating with key strength and weakness
| Study (Author, Year) | Type | AMSTAR-2 Rating | Key Strengths | Key Limitations |
|---|---|---|---|---|
| Liu et al. | Meta-analysis | Moderate to Low | Appropriate methods, duplicate review, RoB assessed | No protocol, no study details, no heterogeneity discussion |
| Slagter et al. | Systematic Review + Meta-analysis | Moderate | Protocol registered, RoB and publication bias assessed | No excluded studies list, partial statistical reporting |
| Madhavan et al. | Systematic Review + Meta-analysis | Moderate | Strong methods, heterogeneity explained, RoB assessed | No RoB impact discussion, no excluded studies listed |
| Wagenaar et al. | Systematic Review | Low | Search strategy, RoB assessed, duplicate review | No protocol, no PICO, no CoI, no heterogeneity or publication bias |
| O’Connor-Cordova et al. | Systematic Review + Meta-analysis | Moderate to Low | Protocol, bias handling, meta-analysis methods | No heterogeneity or publication bias assessment, weak recommendation |
| Territo et al. | Systematic Review | Low | Clear PICO, good process, RoB assessed | No protocol, no RoB impact, no CoI, no publication bias, no recommendation |
RoB: Risk of Bias; PICO: Population, Intervention, Comparator, Outcome; CoI: Conflict of Interest
Surgical and post-operative outcomes were extracted. A summary of reported effect sizes and heterogeneity (I²) is provided in Table 4. Summary of evidence is also presented in Table 5.
Table 4.
Reported outcome
| Article | Article 1 | Article 2 | Article 3 | Article 4 | Article 5 | Article 6 | |
|---|---|---|---|---|---|---|---|
| Type | Meta-Analysis | Meta-Analysis | Meta-Analysis | Meta-Analysis | Systematic Review | Systematic Review | |
| AMSTAR-2 | Moderate to low | Moderate | Moderate | Moderate to low | Low | Low | |
| Quality of AMSTAR-2 | Critically low | Critically low | Critically low | Low | Critically low | Critically low | |
| WIT | OR/RR/MD | MD : 0.13 min | Hedg : 0.077 | Robot longer | Same | ||
| CI | −0.08/0.35 | −0.125/0.278 | |||||
| I2 | 0 | 55.26 | |||||
| Outcome | Same | Same | |||||
| CIT | OR/RR/MD | MD: 4.78 min | MD : 5.18 min | Hedg: −0.353 | Robot longer (NS) | Robot sightly longer | |
| CI | 1.56/8 | 3.99/6.38 | 0.15/0.555 | ||||
| I2 | 11 | NA | 35.66 | ||||
| Outcome | Robot longer | Robot longer | Robot longer | ||||
| Rewarming time | OR/RR/MD | MD : 20.83 min | NA | Hedg: 0.41 | MD = 11.24 | Robot longer | |
| CI | 14.97/26.69 | NA | 0.237/0.583 | −0.46/22.01 | |||
| I2 | 61 | NA | 59.42 | NA | |||
| Outcome | Robot longer | Robot longer | Robot longer | Robot longer | |||
| Total ischemia time | OR/RR/MD | MD: 17.82 min | MD = 16.51 | Robot longer | |||
| CI | 4.72/30.91 | 9.86/23.16 | |||||
| I2 | 86 | NA | |||||
| Outcome | Robot longer | Robot longer | |||||
| Blood loss | OR/RR/MD | MD = − 16.06 mL | MD = − 54.74 mL | Hedg: −0.398 | MD = −53.68 | Robot better | |
| CI | −35.16/3.04 | −96.6/12.58 | −0.537/−0.259 | −89.78/−17.58 | |||
| I2 | 32 | NA | 19.9 | NA | |||
| Outcome | Same | Robot less | Robot less | Robot less | |||
| Blood transfusion | OR/RR/MD | RR: 0.49 | Robot 29% less | ||||
| CI | 0.23/1.04 | ||||||
| I2 | 0 | ||||||
| Outcome | Same | ||||||
| DGF | OR/RR/MD | RR: 1.10 | RR = 0.99 | RR = 0.927 | RR = 1.23 | Robot better | |
| CI | 0.49/2.44 | 0.63/1.54 | −0.614/1.401 | 0.40/3.74 | |||
| I2 | 0 | NA | 0 | 0 | |||
| Outcome | Same | Same | Same | Same | |||
| SSI | OR/RR/MD | RR: 0.22 | RR = 0.15 | RR = 0.331 | RR = 0.31 | Robot better | Robot better |
| CI | 0.06/0.86 | 0.06/0,38 | 0.131/0.837 | 0.19/0.52 | |||
| I2 | 0 | NA | 26.01 | NA | |||
| Outcome | Robot lower | Robot lower | Robot lower | Robot lower | |||
| Hospital stay | OR/RR/MD | MD: −2.03 d | MD = − 1.69 d | Hedg: −0.088 | MD = −3.07 | Open longer | |
| CI | −5.16/1.11 | −0.15/3.22 | −0.278/0.102 | −6.73/−0.59 | |||
| I2 | 76 | NA | 65 | NA | |||
| Outcome | Same | Same | Same | Open longer | |||
| Graft rejection | OR/RR/MD | RR: 1.16 | RR = 1.06 | RR = 1.215 | RR = 0.96 | ||
| CI | 0.73/1.83 | 0.61/1.84 | 0.718/2.057 | 0.55/1.7 | |||
| I2 | 0 | 2 | 0 | NA | |||
| Outcome | Same | Same | Same (NS) | Same | |||
| Graft failure | OR/RR/MD | RR: 0.94 | Same | ||||
| CI | 0.60/1.48 | ||||||
| I2 | 0 | ||||||
| Outcome | Same | ||||||
| All cause mortality | OR/RR/MD | RR: 1.16 | |||||
| CI | 0.42/3.19 | ||||||
| I2 | 0 | ||||||
| Outcome | Same | ||||||
| Operation time | OR/RR/MD | MD : 24.28 min | Hedg : 0.127 | MD = 6.47 | Robot longer (NS) | Robot longer | |
| CI | 4.06/44.49 | 0.036/0.290 | −12.22/25.16 | ||||
| I2 | NA | 47.4 | NA | ||||
| Outcome | Robot longer | Same | Same | ||||
| Incision length | OR/RR/MD | MD = − 9.13 cm | Hedg: −0.567 | Robot better | |||
| CI | 8.14/10.13 | −0.788/−0.347 | |||||
| I2 | NA | 58.82 | |||||
| Outcome | Robot shorter | Robot shorter | |||||
| Incisional hernia | OR/RR/MD | NA | NA | Robot better | |||
| CI | NA | NA | |||||
| I2 | NA | NA | |||||
| Outcome | NS | Robot better | |||||
| Lymphocele | OR/RR/MD | RR = 0.20 | RR = 0.468 | RR = 0.16 | |||
| CI | 0.04/0.89 | −0.152/1.44 | 0.06/0.43 | ||||
| I2 | NA | 0 | NA | ||||
| Outcome | Robot lower | Robot lower | Robot lower | ||||
| Venous thrombosis | OR/RR/MD | 10 open vs. 0 robot | RR = 0.588 | ||||
| CI | −0.212/1.628 | ||||||
| I2 | 0 | ||||||
| Outcome | Same | ||||||
| Arterial stenosis | OR/RR/MD | 3 open vs. 1 robot | RR = 0.588 | ||||
| CI | −0.212/1.628 | ||||||
| I2 | 0 | ||||||
| Outcome | Same | ||||||
| Ureter leak | OR/RR/MD | 2 open vs. 3 robot | 4 open vs. 1 robot | ||||
| CI | |||||||
| I2 | |||||||
| Outcome | |||||||
| UTI | OR/RR/MD | RR = 1.164 | |||||
| CI | −0.597/2.267 | ||||||
| I2 | 0 | ||||||
| Outcome | Open better | ||||||
| Pain | OR/RR/MD | MD = −1.38 | Hedg: −0.45 | MD = −1.45 | Robot better | ||
| CI | −1.15/−1.61 | −0.578/−0.322 | −1.74/−1.16 | ||||
| I2 | NA | 1.177 | NA | ||||
| Outcome | Robot lower | Robot lower | Robot lower | ||||
| Ileus rate | OR/RR/MD | RR = 3.30 | |||||
| CI | 1.25/8.69 | ||||||
| I2 | NA | ||||||
| Outcome | Robot more (NS) | ||||||
| Arterial anastomosis time | OR/RR/MD | Hedg : 0.167 | MD = 2.33 | Robot longer (non significant) | |||
| CI | −0.395/0.728 | −1.67/6.37 | |||||
| I2 | 88.82 | NA | |||||
| Outcome | Same | Same | |||||
| Venous anastomosis time | OR/RR/MD | Hedg: 0.207 | MD = 2.89 | Robot longer (non significant) | |||
| CI | −0.334/0.748 | −2.94/8.72 | |||||
| I2 | 88 | NA | |||||
| Outcome | Same | Same | |||||
| Ureteral anastomosis time | OR/RR/MD | Hedg: 0.247 | MD = 3.88 | Robot longer (non significant) | |||
| CI | −0.219/0.713 | −4.58/12.33 | |||||
| I2 | 83.91 | NA | |||||
| Outcome | Same | Same | |||||
| Hospital readmission | OR/RR/MD | RR = 1.09 | |||||
| CI | 0.10/11.6 | ||||||
| I2 | NA | ||||||
| Outcome | Same | ||||||
Article 1: Robot-Assisted versus Conventional Open Kidney Transplantation: A Meta-Analysis
Article 2: Robot-assisted Kidney transplantation as a minimally invasive approach for kidney transplant recipients: A systematic review and meta-analyses
Article 3: Comparison of outcomes after open versus robotic Kidney transplantation: A systematic review and meta-analysis
Article 4: Living donor robotic assisted kidney transplant compared to traditional living donor open kidney transplant. Where do we stand now? A systematic review and Meta-analysis
Article 5: Minimally Invasive, Laparoscopic, and Robotic-assisted Techniques Versus Open Techniques for Kidney Transplant Recipients: A Systematic Review
Article 6: Open versus robotic-assisted kidney transplantation: A systematic review by the European Association of Urology (EAU) - Young Academic Urologists (YAU) Kidney Transplant Working Group
WIT: Warm ischemia time; CIT: Cold ischemia time; SSI: Surgical site infections; UTI: Urinary tract infection, OR: Odds ratio, CI: Confidence interval, I2: Heterogeneity test, NS: Non-significant
Table 5.
Evidence summary
| Outcome | Number of reviews | Favored approach | Effect size | I2 | AMSTAR-2 summary |
|---|---|---|---|---|---|
| WIT | 4 | Open in 1, Equal in 3 | MD = 0.13, Hedge = 0.07 |
0% 55.26% |
Moderate, Moderate to low, Low |
| CIT | 5 | Open in all | MD = 4.78, MD = 5.18, Hedge = 0.353 |
11% 35.6% |
Moderate, Moderate to low, Low |
| Rewarming time | 5 | Open in all | MD = 20.83, MD = 11.24, Hedge = 0.41 |
59.42% 61% |
Moderate, Moderate to low, Low |
| Total ischemia time | 3 | Open in all | MD = 17.82, MD = 16.51 | 86% | Moderate to low, Low |
| Blood loss | 5 | Robot in 4, Equal in 1 | MD = −16.06, MD = −54.74, MD = −53.68, Hedge = 0.398 |
19.9% 32% |
Moderate, Moderate to low, Low |
| Blood transfusion | 2 | Robot in 1, Equal in 1 | RR = 0.49 | 0% | Moderate to low |
| Delayed Graft function | 5 | Equal in 4, Robot in 1 | RR = 1.10, RR = 0.99, RR = 0.927, RR = 1.23 | 0% | Moderate, Moderate to low, Low |
| SSI | 6 | Robot in all | RR = 0.22, RR = 0.15, RR = 0.331, RR = 0.31 |
0% 26.01% |
Moderate, Moderate to low, Low |
| Hospital stay | 5 | Robot in 2, Equal in 3 | MD = −2.03, MD = 1.69, MD = 3.07, Hedge = 0.088 |
65% 76% |
Moderate, Moderate to low, Low |
| Graft rejection | 4 | Open in 1, Equal in 3 |
RR = 1.16, RR = 1.06, RR = 1.215, RR = 0.96 |
0% | Moderate, Moderate to low |
| Graft failure | 2 | Equal in all | RR = 0.94 | 0% | Moderate to low, Low |
| All cause mortality | 1 | Equal in all | RR = 1.16 | 0% | Moderate to low |
| Operation time | 5 | Open in 3, Equal in 2 | MD = 24.28, MD = 6.47, Hedge = 0.127 | 47.40% | Moderate, Moderate to low, Low |
| Incision length | 3 | Robot in all | MD = −9.13, Hedge = 0.567 | 58.82% | Moderate, Low |
| Incisional hernia | 3 | Open in 2, Equal in 1 | NA | NA | Moderate, Low |
| Lymphocele | 3 | Robot in all | RR = 0.20, RR = 0.468, RR = 0.16 | 0% | Moderate, Moderate to low |
| Venous thrombosis | 2 | Open in 1, Equal in 1 | RR = 0.588 | 0% | Moderate |
| Arterial stenosis | 2 | Open in 1, Equal in 1 | RR = 0.588 | 0% | Moderate |
| Ureter leak | 2 | Robot in all | NA | NA | Moderate |
| UTI | 1 | Open in all | NA | 0% | Moderate |
| Pain | 4 | Robot in all | MD = −1.38, MD = −1.45, Hedge = −0.45 | 1.18% | Moderate, Moderate to low, Low |
| Ileus rate | 1 | Open in all | RR = 3.30 | NA | Moderate to low |
| Hospital readmission | 1 | Equal in all | RR = 1.09 | NA | Moderate to low |
| Arterial anastomosis time | 2 | Equal in 2, Open in 1 | MD = 2.33, Hedge = 0.167 | 88.82% | Moderate to low, Low |
| Venous anastomosis time | 2 | Equal in 2, Open in 1 | MD = 2.89, Hedge = 0.207 | 88% | Moderate to low, Low |
| Ureteral anastomosis time | 2 | Equal in 2, Open in 2 | MD = 3.88, Hedge = 0.247 | 83.91% | Moderate to low, Low |
WIT: Warm ischemia time; CIT: Cold ischemia time; SSI: Surgical site infections; UTI: Urinary tract infection
Each outcome was evaluated for consistency across studies and graded for strength of evidence based on AMSTAR-2 quality ratings. For example, post-operative pain showed high consistency with moderate to low strength, operative time showed moderate consistency with low strength and blood transfusion showed low consistency with very low strength. Table 6 summarizes consistencies and strengths. Lastly, a risk of bias using the AMSTAR-2 Risk of Bias Assessment tool was generated (Fig. 3).
Table 6.
Consistency and strength of outcomes
| Outcome | Consistency Across Reviews | Quality of Reviews (AMSTAR-2) | Conclusion Strength |
|---|---|---|---|
| WIT | Moderate | Critically low | Low |
| CIT | High | Critically low | Moderate to low |
| Rewarming time | High | Critically low | Moderate to low |
| Total ischemia time | High | Critically low | Moderate to low |
| Blood loss | High | Critically low | Moderate to low |
| Blood transfusion | Low | Critically low | Very low |
| Delayed Graft function | High | Critically low | Moderate to low |
| SSI | High | Critically low | Moderate to low |
| Hospital stay | Moderate | Critically low | Low |
| Graft rejection | Moderate | Critically low | Low |
| Graft failure | High | Critically low | Moderate to low |
| All cause mortality | High | Critically low | Moderate to low |
| Operation time | Moderate | Critically low | Low |
| Incision length | High | Critically low | Moderate to low |
| Incisional hernia | Low | Critically low | Very low |
| Lymphocele | High | Critically low | Moderate to low |
| Venous thrombosis | Low | Critically low | Very low |
| Arterial stenosis | Low | Critically low | Very low |
| Ureter leak | High | Critically low | Moderate to low |
| UTI | High | Critically low | Moderate to low |
| Pain | High | Critically low | Moderate to low |
| Ileus rate | High | Critically low | Moderate to low |
| Hospital readmission | High | Critically low | Moderate to low |
| Arterial anastomosis time | Low | Critically low | Very low |
| Venous anastomosis time | Low | Critically low | Very low |
| Ureteral anastomosis time | Low | Critically low | Very low |
WIT: Warm ischemia time; CIT: Cold ischemia time; SSI: Surgical site infections; UTI: Urinary tract infection
Fig. 3.
AMSTAR-2 Risk of Bias
Discussion
Before interpreting the findings of this umbrella review, it is important to acknowledge that the totality of the synthesized evidence rests on a foundation of non-randomized observational studies — predominantly retrospective cohort comparisons — that have been subsequently pooled in systematic reviews and meta-analyses. To date, no high-quality randomized controlled trial comparing RAKT to OKT has been published, and the inherent limitations of observational data — including selection bias, confounding by indication, and center-volume effects — cannot be fully mitigated by meta-analytic pooling alone. All conclusions drawn in this review should therefore be interpreted within this evidentiary context.
In summary, current evidence of low-to-moderate methodological quality, suggests that RAKT may be associated with advantages in blood loss, incision length, postoperative pain, surgical site infection, and lymphocele rate, at the cost of modestly prolonged cold ischemia time and operative duration. Graft rejection, failure, and delayed graft function appeared comparable between techniques, though these findings require confirmation by high-quality prospective studies before definitive clinical recommendations can be made.
A fundamental source of heterogeneity across the included reviews that deserves explicit acknowledgment is the distinction between living and deceased donor kidney transplantation. RAKT has been adopted predominantly in the living donor setting, where the elective nature of the procedure allows for advance robotic setup, controlled operative timing, and optimized team preparation, conditions that favor the robotic approach and minimize its inherent setup-related time burden [10]. In contrast, deceased donor transplantation introduces logistical constraints, time pressure, and greater recipient comorbidity that may differentially impact robotic versus open surgical performance [9]. Among the six reviews included in this umbrella review, only O’Connor-Cordova et al. restricted inclusion exclusively to living donor transplantation, while the remaining five incorporated mixed populations of living and deceased donor recipients in varying proportions. This heterogeneity in donor type composition has direct implications for the interpretation of several key outcomes. Ischemia times; including cold ischemia time, rewarming time, and total ischemia time — are inherently longer and more variable in deceased donor transplantation, and the consistently observed prolongation of these parameters with RAKT may be disproportionately driven by deceased donor cases where robotic setup time compounds an already extended ischemic period. Similarly, delayed graft function, which is strongly associated with deceased donor transplantation and prolonged cold ischemia, cannot be meaningfully compared across reviews with differing donor type compositions without subgroup analysis. Complication rates, including surgical site infection, lymphocele formation, and incisional outcomes, may also differ between donor settings due to differences in recipient selection, urgency of surgery, and immunological risk profiles. The favorable perioperative outcomes observed with RAKT in this umbrella review; particularly in blood loss, SSI, pain, and incision length, are therefore likely most robust in the living donor setting, where the conditions for robotic surgery are optimal and recipient selection is most favorable. Whether these advantages extend equally to deceased donor transplantation has begun to be addressed by Afferi et al., who conducted the first multicenter propensity score-matched analysis of RAKT versus OKT specifically in the deceased donor setting, including 676 patients from seven European academic centers over a nine-year period. Their findings demonstrated comparable short- and long-term functional outcomes between techniques; including dialysis-free survival, graft survival, reintervention-free survival, and overall survival, with similar overall complication rates. Notably, RAKT was associated with reduced rewarming time and vascular anastomosis time even in the deceased donor context, suggesting that experienced robotic teams can maintain the technical advantages of the robotic platform despite the inherent time constraints of deceased donor surgery [9]. In the living donor setting, the same group demonstrated that RAKT was associated with improved perioperative outcomes and lower morbidity compared to OKT while preserving equivalent long-term graft and patient survival [10]. Taken together, these two propensity-matched analyses provide the strongest currently available evidence that RAKT is a safe alternative to OKT across both donor types, and lend important external validity to the directional findings of this umbrella review. Therefore, future umbrella reviews and primary studies should mandate separate reporting of outcomes by donor type to allow meaningful synthesis and avoid the confounding that has limited the interpretability of the current evidence base.
Our study showed that WIT was the same between both groups, even though average I2 score showed moderate heterogeneity, and the studies included had no statistically significant results for this outcome. In fact, WIT is highly dependent on surgeon experience, therefore, prior experience and the fast learning curve of the robot makes it so it doesn’t affect WIT [12, 13]. Surgeon experience and institutional learning curve represent a second major contributor: as robotic experience accumulates, these differences are expected to diminish, as suggested by comparative data from high-volume centers [8]. Some might argue that WIT in robotic kidney retrieval can be longer due to “lack of space” on the operating table or the need to undock the robot before proceeding with kidney retrieval [14], but our study, limited of strength as it may be, showed no difference between both groups.
Cold ischemia time was longer with robotic surgery through all studies with low heterogeneity between studies and statistically significant results. This could be due to the time needed to set up the robot before tackling the surgery. Moreover, abdominal access for kidney placement in robotic surgery takes more time due to the need to place access ports, proceed with regional dissection to make place for the kidney then proceed with the anastomosis. On the other hand, the overall estimate for CIT difference is only 4.98 min, a negligible time compared to the total CIT.
We can then deduce that the total ischemia time is longer in case of robotic transplant since CIT alone is longer, as found in this paper. This result is statistically significant despite the high heterogeneity (86%).
In transplant surgery, especially in renal transplant surgery, ischemia times are important in terms of graft function. Therefore, in case of prolonged ischemia time, the transplanted kidney may experience a delay in function. Our study showed that delayed graft function (DGF) was similar in both groups with an absence of heterogeneity (0%) but with an absence of statistical significance. The most important determinant for DGF is WIT [15]: since the latter is the same between both group, this could explain the absence of difference found in this study. CIT also plays a role in DGF [15, 16], however, since the difference is only approximately 5 min, and in cadaveric donors CIT can go to up to 20 h [17], the difference is insignificant to cause DGF between both groups.
Adding to that CIT plays a role in graft failure and graft loss. Only Lui et al. analyzed this outcome and showed, without heterogeneity (0%) but absence of statistical significance, that graft failure was the same amongst both groups. It’s important to note that other factors influence graft function and survivability such as donor and recipient related factors, therefore CIT cannot be used alone to interpret graft failure rate. Moreover, studies showed that each hour of CIT increases the risk of graft loss, especially if CIT is higher than 30 h [16], that said, this difference of 5 min wouldn’t cause graft failure.
The most feared complication of renal transplant surgery is graft rejection. Out of four included articles that studied this complication, three showed no difference in rejection rate in both groups with absence of heterogeneity and no statistical significance. Madhavan et al. reported a numerically higher rejection rate in the robotic group (RR: 1.215), however this result was not statistically significant (Heterogeneity = 0%) and therefore cannot be interpreted as favoring either technique. Like graft failure, many factors contribute to graft rejection, two of the most important being HLA compatibility and immunosuppressors. With the development of renal transplant, the mandatory pre-operative cross match between donors and recipient, the improvement of the understanding of the underlying immune responses occurring after transplant and during rejection, the mastery of immunosuppressors and the better management of rejection, especially at an early phase, the rate of graft rejection, independent of the surgical method employed, has drastically decreased. In contemporary practice, the rate of rejection occurring at up to two years after transplant is only around 13.9% [18].
Two of the most important causes of graft loss are arterial stenosis and venous thrombosis.
For venous thrombosis Slagter et al. described 10 cases of venous thrombosis using the open technique however none of the patients undergoing robotic transplant developed venous thrombosis. As for arterial stenosis the authors had 3 cases with open surgery and only one with robotic transplants.
Madhavan et al. also reported those complications (but they didn’t differentiate between them) and found no difference between both groups, without heterogeneity (0%) but also without statistical significance.
Ebru et al. stated that the most common cause of vascular complication is vascular trauma during suture or manipulation [19]. The robot’s precision and enhanced vision might decrease the rate of vascular injury, resulting in smoother anastomosis and offering lighter manipulation of the vessels.
Arterial and vascular anastomosis time were comparable in both groups, with a slight increase in robotic surgery, but high heterogeneity and absence of statistical significance renders the strength of evidence low.
All the studies included were consistent with the fact that rewarming time, or in other words, total vascular anastomosis time, for robotic surgery was higher as compared to open transplant, with statistical significance but high heterogeneity (59 and 61%) amongst studies.
Renal transplant surgery consists mostly of three anastomosis: arterial, venous and ureteral. Therefore, since the duration to complete each anastomosis is higher in robotic surgery, total duration will be higher. Moreover, in robotic surgery, robotic docking and undocking take a non negligible amount of time prolonging operative time. Two of the studies included concluded that there are no difference in operative time: this might be due to the bigger incision in open surgery requiring a longer closing time, however the authors did not discuss the cause of the increased time in open surgery. Three others showed that robotic surgery takes longer to complete. It is worthy to note that most results were not statistically significant (only O’Connor-Cordova et al’s results were significant), with moderate to high heterogeneity.
The substantial heterogeneity observed in anastomotic time measurements — with I² values of 88.8% for arterial, 88.0% for venous, and 83.9% for ureteral anastomosis — warrants careful consideration. Several factors likely contribute to this variability. First, surgeon experience and institutional learning curve represent a major source of heterogeneity: robotic vascular anastomosis requires a distinct skillset from open surgery, and operative times reported from centers early in their RAKT experience are unlikely to be comparable to those from high-volume centers with established robotic transplant programs. As proficiency accumulates, anastomotic times with RAKT are expected to decrease, potentially converging with or surpassing open surgery benchmarks [8]. Second, donor type introduces an additional layer of variability: living donor transplantation allows for elective scheduling, optimized team preparation, and controlled operative conditions, whereas deceased donor transplantation introduces time pressure and logistical constraints that may differentially affect robotic versus open approaches [9, 10]. Since the proportion of living versus deceased donor cases varied across included primary studies, this is a plausible contributor to the observed heterogeneity. Third, the evolution of robotic platforms and instrumentation over the study period — spanning publications from 2017 to 2023 — means that earlier studies may reflect outcomes with older systems, while more recent studies benefit from improved robotic technology and refined surgical techniques. Finally, patient-level factors including BMI, prior abdominal surgery, and vascular anatomy complexity are known to influence anastomotic difficulty and time [20], and their distribution was not standardized across primary studies. Taken together, these factors suggest that the pooled anastomotic time estimates should be interpreted with caution, and that future studies should stratify results by donor type, center volume, and surgical experience to allow more meaningful cross-study comparisons.
Most meta-analyses included in this umbrella review reported lower intraoperative blood loss with RAKT, with moderate-to-low heterogeneity (I² = 20–32%) and statistically significant results in the majority of reporting studies. Mean differences in blood loss ranged from − 16.06 mL (Liu et al.) to − 54.74 mL (Slagter et al.), consistently favoring the robotic approach across independent meta-analyses — a finding that is further supported by a Hedges’ g of − 0.398 (I² = 19.9%) reported by Madhavan et al., indicating a small-to-moderate effect size. Regarding blood transfusion rate, only two reviews reported this outcome, limiting the strength of any conclusion. O’Connor-Cordova et al. reported a 29% reduction in transfusion rate with RAKT, while Liu et al. demonstrated a directionally consistent but non-statistically significant decrease, with complete absence of heterogeneity (I² = 0%). Several mechanistic explanations have been proposed for the observed reduction in blood loss with RAKT. The enhanced precision and tremor filtration afforded by robotic instrumentation may reduce inadvertent tissue trauma and unintended vascular injury during dissection and retraction; factors that contribute to diffuse intraoperative bleeding in open surgery [21]. Additionally, the magnified operative field may facilitate more meticulous vascular anastomosis construction, potentially minimizing anastomotic leakage points and reducing the need for intraoperative hemostatic maneuvers. However, these explanations are mechanistically plausible but remain unconfirmed by direct comparative data from the included reviews, and should be regarded as hypotheses rather than established findings. The consistency of direction across independent meta-analyses with low-to-moderate heterogeneity represents the strongest statistical signal in favor of RAKT in this umbrella review, though the low-to-moderate methodological quality of the underlying evidence and the observational nature of the primary studies necessitate cautious interpretation of this finding pending confirmation by prospective comparative data.
Aside from blood loss advantages, robotic surgery has been known for its minimal incisions, less traction on skin leading to a decrease in post-operative pain thus less use of analgesic and faster recovery rate.
Three papers compared incision length between open and robotic kidney transplant, all of which were in favor of robotic surgery. Results were statistically significant but with substantial heterogeneity (59%), which can be explained by two factors: Differences in surgical technique between centers — particularly the number, size, and placement of robotic port sites versus the length of the open iliac fossa incision — varied across primary studies and were not uniformly reported. Additionally, patient body habitus, particularly BMI, influences incision length in open surgery [20] disproportionately compared to robotic surgery, where port placement remains relatively standardized regardless of abdominal wall thickness.
We could then deduce, smaller incision equals less incisional hernia, which was also found in these studies (however no effect type or confidence interval were available). In open kidney transplant, Simson et al. described incisional hernia rate to be around 1.1 to 7% with a mean of 3.2% [22]. Of the main risk factors of incisional hernia, the authors described that surgical site infection and prolonged surgical time may cause incisional hernias [22]. Moreover, studies have found that Hockey-stick incision, used in open transplant surgery, cause more incisional hernias compared to an oblique incision [22].
All six studies included found that surgical site infection (SSI) was lower in the robotic group. Heterogeneity score was low (mean of 13%) with statistical significance. Literature review is also in favor of robotic surgery in terms of SSI: Hou et al. stipulated that smaller incisions lead to a decrease in area of skin contamination that can decrease SSI [23]. Moreover, high amount of blood loss, especially more than 100 cc, could lead to an increase in SSI [23].
Therefore, since blood loss in robotic surgery was lower compared to open surgery and incisions were smaller, it would make sense that SSI are lower using the robot. However, although all six included reviews reported lower SSI rates with RAKT, this finding must be contextualized within the limitations of the evidence base, which comprises observational studies with inherent susceptibility to selection bias and confounding.
Before the surgery, patient’s concern as mostly related to post-operative pain. Robotic surgery has smaller incisions and lower abdominal wall injury [24]. Chiu et al. said that “robotic arms pivot at the port sites and move/rotate around a fixed remote center-of-motion” : this decreases mechanical injury of the abdominal wall [25]. Likewise, trocart stability on the abdominal wall decrease the degree of trauma induced [26]. Four out of the six papers included in our study compared pain outcome between open and robotic surgery, and are in accord with literature finding. All of them were in favor of robotic surgery with absence of heterogeneity (close to 0%) and statistical significance. The consistent superiority of RAKT in postoperative pain scores across four reviews is encouraging, though the subjective nature of pain assessment and the absence of standardized measurement tools across primary studies limit the strength of this conclusion.
Just like vascular anastomosis time, ureteral anastomosis time was similar between both groups with low strength of evidence due to high heterogeneity and absence of statistical significance.
Ureteral anastomosis should be watertight to avoid a minor, treatable, yet bothersome complication: Ureteral leak.
The primary mechanism underlying ureteral leak following kidney transplantation is distal ureteral ischemia, most commonly resulting from excessive peri-ureteral fat dissection during implantation, which disrupts the delicate periureteral vasculature responsible for distal ureteral perfusion [27]. While two of the included reviews reported on ureteral leak rates, their findings were contradictory; Slagter et al. identified a trend favoring RAKT, whereas Madhavan et al. reported a trend favoring open surgery; and neither reached statistical significance, precluding any definitive conclusion regarding the superiority of either technique in preventing this complication. A plausible hypothesis is that the magnified visualization afforded by the robotic platform, while advantageous for vascular anastomosis, may paradoxically encourage more extensive peri-ureteral dissection than would be performed under direct open vision, potentially increasing the risk of distal ureteral devascularization. However, this remains speculative and is not directly supported by the available evidence. Standardized reporting of ureteral dissection technique and extent across future comparative studies would be necessary to test this hypothesis and determine whether surgical approach meaningfully influences ureteral leak risk in kidney transplantation.
Lymphocele; defined as a lymphatic fluid accumulation surrounded by a pseudo-membrane [28], is a recognized complication of kidney transplantation, most commonly resulting from extensive peri-iliac dissection in the recipient or capsular tears during donor nephrectomy or graft preparation [28]. In this umbrella review, RAKT consistently favored lower lymphocele rates across all three reporting reviews, with absent heterogeneity (I² = 0%) and statistically significant results in two of the three studies, representing one of the most consistent and statistically robust findings in favor of the robotic approach. Several mechanistic explanations have been proposed for this observation. The enhanced magnification and precision of robotic dissection may facilitate more meticulous peri-renal fat handling, potentially reducing the risk of capsular disruption compared to open surgery. Additionally, the magnified operative field may allow earlier identification and more efficient coagulation of small lymphatic vessels that would not be readily visible under direct open vision, theoretically reducing lymphatic leakage at the recipient site. However, these explanations remain hypothetical and are not directly tested or confirmed by any of the included reviews. The absence of heterogeneity across studies strengthens the consistency of the directional finding, but the low-to-moderate methodological quality of the underlying evidence means that this conclusion should be regarded as promising rather than definitive, and warrants confirmation by prospective studies with standardized lymphocele assessment protocols.
Urinary tract infections represent a particularly relevant complication in kidney transplant recipients, who are subject to intensive immunosuppressive regimens that significantly impair host defense mechanisms. Beyond immunosuppression, several additional factors contribute to UTI risk in this population, including prolonged cold ischemia time, extended urinary catheterization duration, and the presence of a ureteral stent [29]. Only one of the six included reviews reported on UTI rates, finding a non-statistically significant trend favoring open surgery with complete absence of heterogeneity (I² = 0%). Given that this finding is derived from a single review without statistical significance, no conclusion regarding the differential effect of RAKT versus OKT on UTI risk can be drawn from the current evidence base. Mechanistic arguments can be constructed in both directions: the enhanced anastomotic precision of robotic surgery may theoretically permit earlier catheter and stent removal, potentially reducing infection risk, while the consistently longer cold ischemia times observed with RAKT may counteract this advantage by increasing susceptibility to infectious complications. These remain speculative hypotheses that are not supported by the available data, and prospective studies with standardized infection surveillance protocols and uniform catheter management policies are needed before any conclusion regarding UTI risk can be attributed to surgical technique.
Post-operative ileus represents a physiologically plausible complication of RAKT that warrants consideration given the fundamental anatomical difference between the two approaches. In conventional open kidney transplantation, the donor kidney is placed extra-peritoneally in the iliac fossa without openinig of the peritoneal membrane, thereby avoiding direct bowel manipulation and minimizing the risk of paralytic ileus. Only one of the six included reviews; O’Connor-Cordova et al., reported on ileus incidence, finding a higher rate in the robotic group. However, this finding is derived from a single review restricted exclusively to living donor recipients, and no confidence interval or heterogeneity metric was available for this outcome, further limiting its interpretability. While the higher risk of ileus with RAKT is anatomically and biologically plausible, this finding was reported in only one study and has not been confirmed by subsequent research. As such, it should be interpreted with caution and not considered representative of the overall comparison between RAKT and OKT. This outcome deserves dedicated prospective reporting in future comparative studies, with particular attention to perioperative bowel management protocols and the potential mitigating role of minimizing intraperitoneal insufflation time.
Overall, the available evidence suggests that robotic surgery offers advantages regarding blood loss, SSI, incision length, ureteral leak, and post-operative pain, whereas open surgery appears to perform better in terms of CIT, rewarming time, UTI rate, and ileus rate.
While robotic surgery appears to reduce several perioperative complications, this did not translate into consistent differences in hospital stay, all-cause mortality, or hospital readmission, outcomes that are multifactorial and likely influenced by factors beyond surgical technique alone.
Hospital stay demonstrated the most inconsistent findings among all reported outcomes, with substantial-to-high heterogeneity (I² = 65–76%) and no clear directional consensus across the five reporting reviews. Two reviews reported a modest reduction in hospital stay favoring RAKT (MD = − 2.03 days and MD = − 3.07 days respectively), while three reviews found no significant difference between techniques. This variability is unsurprising given that length of hospital stay is among the most multifactorial and institutionally dependent outcomes in surgical research, influenced by factors that extend well beyond the surgical technique itself. Donor type is a primary contributor: living donor recipients generally experience shorter and more predictable postoperative courses than deceased donor recipients, who are more susceptible to delayed graft function and its associated prolonged hospitalization [9, 10]. Since the proportion of living versus deceased donor cases varied substantially across included primary studies — and one review (O’Connor-Cordova et al.) studied exclusively living donor recipients — direct cross-study comparisons of hospital stay are inherently limited. Institutional discharge protocols represent an equally important source of variability: criteria for safe discharge differ markedly between centers and healthcare systems, meaning that a clinically similar patient may be discharged on day three at one center and day seven at another, independent of surgical approach. Furthermore, surgeon experience and the learning curve of RAKT may prolong postoperative monitoring requirements at centers early in their robotic transplant program, artificially inflating hospital stay in the robotic arm of earlier studies. Patient-level factors including BMI, comorbidity burden, and immunosuppression tolerance further compound this variability. Taken together, the available evidence does not support a definitive conclusion regarding the effect of surgical technique on hospital stay, and future studies should prospectively standardize discharge criteria and stratify analyses by donor type and center experience to allow meaningful comparison of this outcome.
Even though complications occur, all-cause mortality is generally related to the patient’s risk factors and comorbidities. Patient selection between both groups should be almost comparable, making mortality similar in both groups, as found in this review. Only Lui et al. mentioned this with no heterogeneity (0%) but their results were not statistically significant.
Likewise, hospital readmission is multifactorial, including surgical and non-surgical complications such as transplant, creatinine elevation, immunosuppressors adverse effects, … Therefore, it would be logical that the rate of readmission is similar between both groups (even though only O’Connor-Cordova et al. mentioned this complication and results weren’t statistically significant).
Hospital readmission following kidney transplantation is a multifactorial outcome influenced by a broad spectrum of surgical and non-surgical factors, including wound complications, graft dysfunction, creatinine elevation, immunosuppressor-related adverse effects, and infectious complications. Only one of the six included reviews; O’Connor-Cordova et al., reported on hospital readmission rates, finding no statistically significant difference between RAKT and OKT (RR = 1.09). Given that this represents a single non-replicated observation from a study restricted exclusively to living donor recipients, no conclusion regarding the differential effect of surgical technique on hospital readmission can be supported by the current evidence. The multifactorial nature of hospital readmission means that any genuine technique-related differences, if they exist, would likely be overlooked by the dominant influence of immunological, infectious, and patient-level factors that are independent of surgical approach. Prospective studies reporting readmission rates with standardized definitions, stratified by donor type and readmission cause, are necessary before meaningful conclusions can be drawn regarding the impact of surgical technique on this outcome.
Cost remains a critical and frequently underappreciated dimension of the RAKT versus OKT comparison, particularly in the context of global healthcare resource allocation and the expanding adoption of robotic platforms in transplant surgery. Among the six reviews included in this umbrella review, only Slagter et al. reported on cost outcomes, identifying three primary studies with available cost data. Oberholzer et al.; the only study to formally analyze cost differences, reported a significantly higher total cost per transplant for RAKT compared to OKT ($75,118 versus $60,552 respectively; p = 0.02), representing an approximate 24% cost premium for the robotic approach. Two additional studies reported more modest additional costs attributable to RAKT-specific consumables, ranging from $575 for a dedicated access device (Ahlawat et al.) to $3,000 in additional procedural costs (Maheshwari et al.), though neither provided full cost breakdowns. These figures reflect not only the per-procedure instrument costs but also the broader financial burden of robotic surgery, including the initial capital investment in robotic platforms and ongoing maintenance contracts. However, a narrow focus on direct procedural cost risks underestimating the potential downstream economic benefits of RAKT. Reductions in surgical site infection rates, lymphocele formation, postoperative pain, and incision-related complications; all consistently observed in favor of RAKT in this umbrella review, carry their own cost implications, including reduced antibiotic utilization, fewer interventional procedures for complication management, and potentially shorter active recovery periods. Slagter et al. themselves acknowledged that a shorter hospital stay and decreased SSI and lymphocele rates could partially lead to higher procedural costs, and that total costs of robotic surgical procedures do not necessarily exceed those of open surgery when downstream savings are factored in. A comprehensive health economic evaluation of RAKT versus OKT would therefore require a full cost-effectiveness analysis incorporating direct procedural costs, complication-related expenditure, readmission rates, and long-term graft function outcomes. Furthermore, as new robotic platforms enter the market and manufacturer competition intensifies, the per-procedure cost differential between RAKT and OKT is likely to narrow over time, as has been observed in other robotic surgical specialties. Until formal cost-effectiveness analyses are conducted using prospective multicenter data, cost alone should not be regarded as a definitive argument against the adoption of RAKT in centers with established robotic infrastructure and sufficient surgical volume to justify the investment.
This is one of the first umbrella study comparing the outcome of renal transplant using the standardized open approach compared to the robot-assisted technique. The study compared most of the outcomes of both procedures, covering most of them, from surgical complications to post-operative outcomes and graft survivability. Moreover, standardized tools were used to build this study such as the PRISMA, the AMSTAR-2, consistency rating and AMSTAR-2 risk of bias.
However, most of the studies included were of moderate to low AMSTAR-2 quality, none of which were of high AMSTAR-2 rating. Also, the quality of evidence was found to be critically low due to heterogeneity of outcomes and weak AMSTAR-2 rating, affecting the strength of the evidence. Details of strength and weaknesses are detailed in Table 3.
This umbrella review was not prospectively registered in a public registry such as PROSPERO prior to commencement, which represents a methodological limitation as it precludes independent verification that outcomes were pre-specified rather than selectively reported after data inspection. Future umbrella reviews on this topic should consider prospective registration to enhance transparency and reproducibility.
A formal CCA analysis revealed very high overlap (CCA = 24.4%) among the included reviews, with several primary studies appearing in up to five of the six reviews. This represents an inherent limitation of umbrella reviews in emerging fields where the primary study pool remains limited, and conclusions should be interpreted accordingly.
It’s also important to note that some outcomes were underreported such as vascular anastomosis failure, venous thrombosis and arterial stenosis, ureteral leak and blood transfusion rate. Lastly, not all studies used the same effect type (MD versus Hedge score) making direct comparison difficult.
Conclusion
In summary, despite these limitations, the study can safely suggest that robotic kidney transplant surgery yields better outcomes in term of blood loss, post-operative pain, incision length and surgical site infection. Same rejection rates, graft failure and mortality rate were found between both techniques. Hence RAKT may improve some operative outcomes while maintaining same graft survival rates. However, the strength of evidence is limited by low methodological quality. Therefore, high quality studies are required to obtain more robust data and strengthen the quality of evidence.
Author contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Roy Madi and Elie Helou. The first draft of the manuscript was written by Roy Madi and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Ethical approval
This study being a review, ethics approval was not necessary.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.GBD Chronic Kidney Disease Collaboration (2020) Global, regional, and national burden of chronic kidney disease, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Lond Engl 395(10225):709–733. 10.1016/S0140-6736(20)30045-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hatzinger M, Stastny M, Grützmacher P, Sohn M (2016) [The history of kidney transplantation]. Urologe A 55(10):1353–1359. 10.1007/s00120-016-0205-3 [DOI] [PubMed] [Google Scholar]
- 3.Axelrod DA et al (2018) An economic assessment of contemporary kidney transplant practice. Am J Transplant 18(5):1168–1176. 10.1111/ajt.14702 [DOI] [PubMed] [Google Scholar]
- 4.Poggio ED, Augustine JJ, Arrigain S, Brennan DC, Schold JD (2021) Long-term kidney transplant graft survival—making progress when most needed. Am J Transplant 21(8):2824–2832. 10.1111/ajt.16463 [DOI] [PubMed] [Google Scholar]
- 5.Jha V, Wang A-M, Wang H (2012) The impact of CKD identification in large countries: the burden of illness. Nephrol Dial Transplant 27(suppl 3):iii32–iii38. 10.1093/ndt/gfs113 [DOI] [PubMed] [Google Scholar]
- 6.Williamson T, Song S-E (2022) Robotic surgery techniques to improve traditional laparoscopy. JSLS J Soc Laparoendosc Robot Surg 26(2):e2022.00002. 10.4293/JSLS.2022.00002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hoznek A et al (2002) Robotic assisted kidney transplantation: an initial experience. J. Urol. 167(4):1604–1606 [PubMed]
- 8.Gallioli A et al (2020) Learning curve in robot-assisted kidney transplantation: results from the European Robotic Urological Society Working Group. Eur Urol 78(2):239–247. 10.1016/j.eururo.2019.12.008 [DOI] [PubMed] [Google Scholar]
- 9.Afferi L et al (2025) Robot-assisted vs open kidney transplantation from deceased donors. BJU Int 136(3):484–492. 10.1111/bju.16799 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Afferi L et al (2026) Robot-assisted versus open kidney transplantation from living donor. Eur Urol Focus 12(2):256–265. 10.1016/j.euf.2025.12.009 [DOI] [PubMed] [Google Scholar]
- 11.Fusar-Poli P, Radua J (2018) Ten simple rules for conducting umbrella reviews. Evid Based Ment Health. 10.1136/ebmental-2018-300014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chahal B et al (2037) The learning curves of major laparoscopic and robotic procedures in urology: a systematic review. Int J Surg 109(7):2037. 10.1097/JS9.0000000000000345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Passerotti CC et al (2015) Comparison of the learning curves and frustration level in performing laparoscopic and robotic training skills by experts and novices. Int Urol Nephrol 47(7):1075–1084. 10.1007/s11255-015-0991-3 [DOI] [PubMed] [Google Scholar]
- 14.Kawan F, Theil G, Fornara P (2018) Robotic donor nephrectomy: against. Eur Urol Focus 4(2):142–143. 10.1016/j.euf.2018.07.027 [DOI] [PubMed] [Google Scholar]
- 15.Bahl D, Haddad Z, Datoo A, Qazi YA (2019) Delayed graft function in kidney transplantation. Curr Opin Organ Transplant 24(1):82–86. 10.1097/MOT.0000000000000604 [DOI] [PubMed] [Google Scholar]
- 16.Liu L, Cheng K, Huang J (2021) Effect of long cold ischemia time of kidneys from aged donors on prognosis of kidney transplantation. Ann Transplant 26:0–0. 10.12659/AOT.928735 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Peters-Sengers H et al (2019) Impact of cold ischemia time on outcomes of deceased donor kidney transplantation: an analysis of a national registry. Transplant Direct 5(5):e448. 10.1097/TXD.0000000000000888 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Alsulami MM et al (2024) The predictors and risk factors of 2-year rejection in renal transplant patients: a multicenter case-control study. Am J Nephrol 55(4):487–498. 10.1159/000538963 [DOI] [PubMed] [Google Scholar]
- 19.Emiroglu R, Karakayali H, Sevmis S, Arslan G, Haberal M (2001) Vascular complications in renal transplantation. Transplant Proc 33(5):2685–2686. 10.1016/S0041-1345(01)02148-0 [DOI] [PubMed] [Google Scholar]
- 20.Dobrzycka M, Bzoma B, Bieniaszewski K, Dębska-Ślizień A, Kobiela J (2022) Pretransplant BMI significantly affects perioperative course and graft survival after kidney transplantation: a retrospective analysis. J Clin Med 11(15):4393. 10.3390/jcm11154393 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tan A et al (2016) Robotic surgery: disruptive innovation or unfulfilled promise? A systematic review and meta-analysis of the first 30 years. Surg Endosc 30(10):4330–4352. 10.1007/s00464-016-4752-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Simson N, Parker S, Stonier T, Halligan S, Windsor A (2018) Incisional Hernia in Renal Transplant Recipients: A Systematic Review. Am. Surg. 84(6):930–937 [PubMed]
- 23.Hou Y, Hu Y, Song W, Zhang J, Luo Q, Zhou Q (2022) Surgical site infection following minimally invasive lobectomy: is robotic surgery superior? Cancer Med 11(11):2233–2243. 10.1002/cam4.4609 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Jin S-J et al (2017) Comparison of postoperative pain between laparoscopic and robot-assisted partial nephrectomies for renal tumors: A propensity score matching analysis. Med (Baltim) 96(29):e7581. 10.1097/MD.0000000000007581 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Chiu L-H, Chen C-H, Tu P-C, Chang C-W, Yen Y-K, Liu W-M (2015) Comparison of robotic surgery and laparoscopy to perform total hysterectomy with pelvic adhesions or large uterus. J Minimal Access Surg 11(1):87–93. 10.4103/0972-9941.147718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Martino MA et al (2011) A cost analysis of postoperative management in endometrial cancer patients treated by robotics versus laparoscopic approach. Gynecol Oncol 123(3):528–531. 10.1016/j.ygyno.2011.08.021 [DOI] [PubMed] [Google Scholar]
- 27.Di Carlo HN, Darras FS (2015) Urologic considerations and complications in kidney transplant recipients. Adv Chronic Kidney Dis 22(4):306–311. 10.1053/j.ackd.2015.04.003 [DOI] [PubMed] [Google Scholar]
- 28.Guachetá-Bomba PL, Sandoval Guerrero MF, Ramirez G, Garcia-Perdomo HA (2023) Lymphocele Complication After Kidney Transplant: Current Literature Review and Management Algorithm. Exp. Clin. Transplant. Off. J. Middle East Soc. Organ Transplant. 21(11):855–859. 10.6002/ect.2023.0037 [DOI] [PubMed]
- 29.Goodfellow M, Thompson ER, Tingle SJ, Wilson C (2023) Early versus late removal of urinary catheter after kidney transplantation. Cochrane Database Syst Rev 2023(7):CD013788. 10.1002/14651858.CD013788.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No datasets were generated or analysed during the current study.